The ever increasing usefulness of the personal computer (PC) leads users to an ever increasing dependence upon its capabilities. Ironically, that dependence in turn highlights some shortcomings of the traditional, “location-based,” model of computing. In that model, users come to a powerful, self-contained PC when they have computing tasks to perform. This model has long been accepted, but clearly preferable is a “user-based” model of computing, wherein computing power comes to the user whenever, and wherever, it is needed.
The proliferation of laptop computers, pocket PCs, and their ilk illustrates a first step toward user-based computing. Users carry computing power with them so that it is always available. Useful as they are, however, portable devices have not eliminated the fixed location PC. The fixed PC coffers greater levels of performance than do many portables. Also, many peripheral devices are best left in one place, whether because of their weight, such as in the case of high-quality printers and scanners, or because of their connection to inherently fixed resources, such as Internet access devices that use DSL, fiber optic, CATV, or other “tethered” communications media. Even if the functionality of these peripherals could be made portable, their power drain would exacerbate an already sore point with users of portable devices. Many users are also hard pressed to justify the expense of owning both fixed and portable versions of some devices, especially as the fixed devices stand idle while the user moves about.
Microsoft's “WINDOWS TERMINAL SERVICES” (WTS) takes another step toward user-based computing, a step beyond laptops. WTS leverages the power of a fixed location PC, making it serve as host for numerous client workstations. The workstations communicate with the host, and each workstation uses the computing power, storage, and other resources of the host to present a user with the experience of running a full PC. Thus, the host can be placed in a location best suited for connecting to its peripherals and to tethered data channels while the workstations are placed at locations more convenient to the users. The host can be always running so that when a user accesses a workstation, he is faced only with a minimal login delay rather than a full boot delay. Because the host supports multiple, simultaneous users, a family that needs another PC may purchase a relatively inexpensive workstation rather than a second full PC. Upgrade costs are reduced as they are localized to the host whose upgraded resources are then made available for use by all users via the client workstations.
However, a PC (or PC-like workstation) in every room does not satisfy a user when his desire is for computing portability, not just computing ubiquity. Portability is especially important when the user wants to use the services of the computer while he is doing something else. For example, he may read news from an Internet service while eating breakfast or may scan sports statistics while watching television. What is needed is a device that separates the parts of the PC that have some reason for staying fixed (either by inherent necessity or by economic desirability) from the parts of the PC that the user needs to access as he wanders. An ideal device would leverage the power and connectivity of the fixed PC while allowing the flexibility of movement and location of portable devices.